Increasingly, Veterans are prescribed potent opioid analgesics for the treatment of chronic pain despite limited evidence for efficacy and increasing evidence of serious harms including addiction and non-fatal and fatal overdose. While guidelines recommend consideration of dose increase for patients not benefitting from opioid therapy, the rates of major harms are directly related to dose. Higher doses may also be more likely to precipitate opioid-induced hyperalgesia, a paradoxical increased pain response, in susceptible individuals. In summary, opioid dose increase, a currently accepted clinical response to poorly controlled pain, may offer little benefit and certainly increases risk, especially in patients already on moderate-high doses (90-120 mg daily morphine equivalents). Alternative treatment strategies to opioid dose escalation that lessen risk and possibly increase benefit are much needed. Switching to buprenorphine/naloxone (BUP/NX), a partial opioid agonist approved for use in the treatment of opioid abuse/dependence, may be a safe and effective alternative strategy to opioid dose escalation in the treatment of chronic pain. As a partial agonist, there is a ceiling to BUP/NX's respiratory depressant and other opioid-like effects, meaning it is less likely to cause addiction and overdose. Additionally, there are pre-clinical data to suggest BUP/NX is less likely to produce opioid-induced hyperalgesia and may even reverse it in patients switched from full agonist opioids. Case series have demonstrated improvements in pain, functional status and quality of life among patients switched from full agonist opioids to BUP/NX for chronic pain. Controlled trials are needed to establish BUP/NX's efficacy compared to opioid dose escalation in the treatment of poorly-controlled pain. We propose a pilot 12-week, open label randomized trial of BUP/NX compared to opioid dose escalation among patients with poorly-controlled pain on the primary outcome of pain intensity. As patient acceptance of either opioid dose escalation or BUP/NX is unknown, our first objective is to assess willingness to enroll in a randomized trial and reasons for and against enrollment among eligible patients. The study will compare treatments on the primary outcome of pain intensity, measured using the 11-point pain numerical rating scale, and secondary outcomes of pain interference, using the Brief Pain Inventory functional interference subscale, medication adherence and patient global assessment of change. Mixed models will be employed in the analysis to accommodate potential unbalanced repeated measures with missing data. Effect size estimates will be used to generate sample size projections for a definitive trial. This line of research is a direct extensio of the PI's HSR&D-funded CDA-2 project developing a screening tool to identify low efficacy opioid use in primary care and also well-aligned with the Strategic Plan and Focused Area of Research of the Pain Research, Informatics, Medical comorbidities, and Education (PRIME) Center's proposal for a Center of Innovation (COIN) and its strategic objective to Promote access, continuity, and sustainability of safe and effective interventions for pain and pain-relate disability.